Precedent on High: The High Court of Australia and ‘Seriously Considered Dicta’

By James Lee

This post considers a specific feature of the High Court of Australia’s approach to precedent. Since the 2007 decision in Farah Constructions v Say-Dee [2007] HCA 22, the Justices have moved to restrict the scope of the lower courts to develop the law, by asserting that lower courts are bound by ‘seriously considered dicta’ ([134] and [158]) of the High Court, admonishing courts below for raising new arguments in the course of judgments. Keith Mason has claimed that, in so doing, the High Court has effected a ‘profound shift in the rules of judicial engagement’ (Keith Mason, ‘President Mason’s Farewell Speech’ (2008) 82 Australian Law Journal 768, 769, see the original remarks here at 18–22). An excellent feature-length examination of the issues has been provided by Associate Professor Matthew Harding and Professor Ian Malkin (‘The High Court of Australia’s Obiter Dicta and Decision-Making in Lower Courts’ (2012) 34 Sydney Law Review 239). In this short post, I argue that the approach has uncertain implications for the Australian doctrine of precedent and the scope for intermediate courts of appeal to develop the law.

What is the High Court’s approach? Farah v Say-Dee and subsequent cases
Farah v Say-Dee concerned a claim for a variety of equitable reliefs in respect of various properties which were the subject of a joint venture development scheme between the claimant and defendant. The planning application for the property was unsuccessful, because the site was considered too small to ‘maximise its development potential’. During the application process, the defendant learned that permission was more likely to be granted if adjacent properties were included in the planned development. The defendant bought these properties through a company which he controlled. The claimants contended that these properties had been acquired through a breach of fiduciary duty and that consequently the recipients had knowingly received the properties, which were held on constructive trust for the claimants. The Court of Appeal of the Supreme Court of New South Wales allowed the claimant’s claim, reversing the judge’s finding that there had been no breach of fiduciary duty, and instead held that the defendants were liable in knowing receipt. The Court of Appeal also found that a strict liability claim in unjust enrichment was available.

The High Court unanimously reversed the judgment of the Court of Appeal, allowing the appeal in emphatic terms. The Court restored the trial judge’s findings and held that there had been no breach of fiduciary in the first place, because Say-Dee had given informed consent to the purchase of the properties. Furthermore, their Honours reprimanded the Court of Appeal (at [131]) for a ‘grave error’, both from the doctrinal view of the relevant principles but also (at [158]) because they ‘were arrived at without notice to the parties, were unsupported by authority and flew in the face of seriously considered dicta uttered by a majority of this Court’. (But apart from that…). Where the High Court has ‘seriously considered’ (at [134] and [158]) an issue, even if it was not necessary for the decision of the case and may not therefore amount to the ratio decidendi, ‘all other Australian courts are bound accordingly’ (Bofinger v Kingsway Group Ltd [2009] HCA 44, [86]). But what does it mean to say that certain observations were ‘seriously considered’?

Justice Heydon has sought to explore this new doctrine in an article where he draws attention to the controversy and uncertainty of the distinction between ratio decidendi and obiter dicta (Justice J D Heydon, ‘How Far Can Trial Courts and Intermediate Appellate Courts Develop the Law?’ (2009) 9 Oxford University Commonwealth Law Journal 1). The High Court has felt it necessary to extend the doctrine of precedent, so that the common practice of the lower courts, in both England and Australia, of following non-ratio dicta becomes compulsory for the Australian courts. On a straightforward reading of the approach, one might think that it is really just confirming existing practice, that everything the High Court says is important, but it seems that there is more to it than that.

The Farah doctrine of the binding force of ‘seriously considered dicta’ has been mentioned in two subsequent High Court cases. In R v Keenan [2009] HCA 1, Justice Kirby, under the heading ‘Obedience to the Court’s authority’, doubted the extent to which lower courts may be bound beyond the ratio of a High Court decision (at [35]):

In recent years, this Court has repeatedly reminded judges at trial and intermediate courts of their duty to conform to the rulings of this Court in matters submitted to it for its decision. It has instructed them to observe ‘seriously considered dicta uttered by a majority of this Court’. Although, respectfully, I question whether the legal duty of obedience extends beyond obedience to the rationes decidendi of earlier decisions, I certainly agree that, where such decisions exist, the legal principles for which they stand must be applied by judicial officers subject to this Court’s authority as an aspect of the rule of obedience to the doctrine of judicial precedent that applies throughout the Judicature of this country.

In Pape v Commissioner of Taxation [2009] HCA 23, a case on the constitutionality of tax legislation, Justice Heydon appeared to add a further qualification (albeit dissenting as to the outcome of the case). The defendants, his Honour said, had relied on ‘statements… [which were] not part of the ratio decidendi of any of the decisions in which they were made. They were seriously considered dicta, but they could not be described as conforming with long-established authority’ (at [473]). Thus, for the Farah doctrine to apply, it seems that dicta must also conform to long-established authority and have been applied as such (see, eg, Sun North Investments Pty Ltd as Trustee v Dale [2013] QSC 44, [114] (Justice Henry)). The Farah doctrine was the subject of debate during the hearing in PGA v The Queen [2011] HCATrans 267, a marital rape case, as to whether the absence of reference to a relevant authority undermined reliance on relevant dicta. M G Hinton QC, Solicitor-General for South Australia, made express reference both to Farah and to Justice Heydon’s article.

The proper role of an intermediate court of appeal has become a motif for extra-judicial observation on several occasions by the Justices of the High Court. In addition to Justice Heydon’s article, Justice Hayne considered the predicament of the judge in a lower court seeking to develop the law:

Faithful application of precedent is at the heart of the judicial task. The justice which a judge must do, is justice according to law. The judge, particularly the judge at first instance, is not free to recast the law at will, whatever he or she may think of it. As Samuel Johnson said:

To permit a law to be modified at discretion is to leave the community without law. It is to withdraw the direction of that public wisdom by which the deficiencies of private understanding are to be supplied. (emphasis added).

From time to time the High Court finds it necessary to point out this fact.

(Justice Kenneth Hayne, ‘Letting Justice Be Done Without the Heavens Falling’ (2001) 27 Monash University Law Review 12, 17–18).

What do the lower courts make of all this?
It is not yet clear how much impact this line of authority will have on Australian jurisprudence. In the six years since Farah, the seriously considered dicta requirement has been expressly considered only intermittently in the lower courts, and the High Court itself has not expressly explored the doctrine since 2009 (in Pape). The High Court’s guidance may well have had considerable influence on the lower courts’ practices, even if not expressly cited. But there is some evidence that the warning against raising new arguments has been taken on board.

In Day v The Ocean Beach Hotel Shellharbour Pty Ltd [2013] NSWCA 250, Justice Leeming (at [31]) considered various observations on vicarious liability from the High Court in Oceanic Crest Shipping Co v Pilbara Harbour Services Pty Ltd [1986] HCA 34 to determine whether they were part of the ratio:

On my reading, those passages demonstrate that common to one of the lines of reasoning employed by the majority in Oceanic Crest was the proposition of law that once one person was vicariously liable, no other person could be. But even if I be wrong about the proposition forming part of the ratio of the case, it amounts to ‘long-established authority and seriously considered dicta’ of the High Court which should be followed.

There is also uncertainty as to how far the seriously considered dicta doctrine extends with regard to dicta from other courts In Director of Public Prosecutions (Vic) v Patrick Stevedores Holdings Pty Ltd [2012] VSCA 300, the Court of Appeal of the Supreme Court of Victoria (President Maxwell, Justices Weinberg and Ferguson) allowed an appeal from the judgment of Justice Robson ([2012] VSC 31) in which his Honour had followed what he viewed as binding ‘seriously considered dicta’ from the Full Court of the Federal Court of Australia in Oates v Williams [1998] FCA 775. The Court of Appeal held, however, that the judge had erred:

The decisions of intermediate appellate courts do not have the precedential weight of decisions of the High Court. Whatever may be the full implications of Farah, and its injunction that ‘seriously considered dicta uttered by a majority’ of the High Court should be regarded as binding, there is nothing to suggest that this principle applies to such dicta in the judgments of intermediate appellate courts.

In our opinion, an intermediate appellate court such as our own, faced with conflicting decisions of other intermediate appellate courts, is not bound to follow any one of those decisions. The position might be different if, post Farah, an intermediate appellate court had said that an earlier decision of that court, or of another intermediate appellate court, was plainly wrong. In such circumstances, this Court might be bound to follow the later decision unless we took the view that the later decision was ‘plainly wrong’. Failing that, this Court is at liberty to state the law as it thinks appropriate.

This decision touches upon other areas of uncertainty in precedent, upon which the High Court’s guidance would be welcomed, such as the relation between intermediate appellate courts across the different states of Australia (see, eg, the High Court’s approval of the Court of Appeal of Western Australia in Newcrest Mining Ltd v Thornton [2012] HCA 60; see Hili v The Queen [2010] HCA 45, [57]; CAL No 14 Pty Ltd v Motor Accidents Insurance Board [2009] HCA 47, [50]). Similarly, Justice Heydon has reiterated (as in Farah) that ‘[a] court may not decide a case on a point not raised by one of the parties or by the court for the consideration of the parties’ (as part of the majority in International Finance Trust Co Ltd v New South Wales Crime Commission [2009] HCA 49, [146]) but it is not clear quite how far this prohibition extends. And indeed Farah itself must be understood as part of a narrative of controversy over the scope and role of the law of unjust enrichment in Australia (as to which, see Andrew Burrows, ‘The Australian Law of Restitution: Has the High Court Lost Its Way?’ in Elise Bant and Matthew Harding (eds), Exploring Private Law (Cambridge University Press, 2010) 67). All those points, are, however, beyond the scope of this present post.

What are the problems?
With respect to the High Court, it cannot be that it is only acceptable for a court of appeal to develop the law when it feels able to second guess the superior court. The case-loads of such courts are radically different, and there are a various factors which will affect whether a case even reaches the final court of appeal, not least the incidents and accidents of litigation.

Even if one may disagree with particular examples of erroneous reasoning or conclusions or both in a court of appeal, the law cannot be allowed to stagnate under a controlling hand of an observation from a judge in the highest court. The High Court of Australia’s position is thus to be regretted, and we may conclude with Justice Paul Finn that ‘the Australian way is not to be one in which the potential for diversity in ideas below the High Court is to be fostered’ (Paul Finn, ‘Internationalisation or Isolation: The Australian Cul de Sac? The Case of Contract Law’ Elise Bant and Matthew Harding (eds), Exploring Private Law (Cambridge University Press, 2010) 41).

The idea of ‘seriously considered dicta’ would appear to imply that some judicial observations are not to be taken seriously (ioculariter dicta, perhaps?). How are lower courts to know which dicta ‘count’? Though I do not agree with such a strict demand of precedent, one possible solution for Justice Heydon could be to argue that lower courts are bound by all dicta of the superior appellate court, unless the judge expressly identifies a passage as not a necessary part of their decision, admitting that they are only speculating. Such an approach would be quite restrictive, but is tenable and seems in line with Justice Heydon’s view. It would, however, require considerable discipline on the part of judges in the High Court not to over-extend themselves in their judgments. It is also questionable whether applying such a new doctrine retrospectively to High Court judgments delivered before the rule was in place is sensible.

Yet we may also wonder, if dicta are not ‘seriously considered’, why have they been put into a judgment? One is reminded of the possibly apocryphal comment (inconsistently attributed to one of two great British managers, Bill Shankly and Brian Clough) about the offside rule in football (‘soccer’, rather than the AFL or NRL varieties). One way in which the offside rule is infringed is if a player is ‘interfering with play’ from an offside position (see here). But, as the line goes, ‘If the player’s “not interfering with play”, what is he doing on the pitch?’

It has been over a quarter of a century since the Australia Act 1986 (Cth), which finally (via s 11) abolished all appeals to the Privy Council from decisions of Australian courts, and Australians became ‘the masters of [their] own legal destiny’ (Sir Anthony Mason ‘Future Directions in Australian Law’ (1987) 13 Monash University Law Review 149, 151). In that time, the High Court of Australia has insisted upon ‘a unified common law which applies in each state but is not itself the creature of any State’ (Kable v Director of Public Prosecutions (NSW) [1996] HCA 24 (Justice McHugh)). The question is how to police the unity, without shutting off ‘much of the oxygen of fresh ideas that would otherwise compete for acceptance in the free market of Australian jurisprudence’ (Keith Mason ‘Farewell Speech’). For example, in Qantas Airways Ltd v Transport Workers’ Union of Australia [2011] FCA 470, Justice Moore of the Federal Court felt constrained from recognising an Australian tort of unlawful interference with trade, given the recent guidance on precedent: [422]–[430].

It is of course legitimate and necessary for a final court of appeal to correct doctrinal or procedural errors on the part of lower courts. But it is crucial to distinguish between the two. The obsequious adherence to every observation of a higher court is not to be favoured: even if it were, it would behove the High Court to explain its criteria for the engagement of its novel, restrictive doctrine of precedent much more clearly, as Harding and Malkin have urged. I have argued above that the Justices have not yet offered such an explanation. Doing so may also compel the Justices to reflect upon their own approach to judging. With respect, the High Court’s approach to the doctrine of precedent urgently needs to be clarified and, perhaps, seriously reconsidered.

AGLC3 Citation: James Lee, ‘Precedent on High: The High Court of Australia and “Seriously Considered Dicta”’ on Opinions on High (21 August 2013) <>.

James Lee is a Lecturer in Law (Senior Lecturer from October 2013) at the University of Birmingham and Academic Fellow of the Honourable Society of the Inner Temple. Both the Universities of Melbourne and Birmingham are members of Universitas 21. His research is principally in private law and on judicial reasoning in superior appellate courts. He was a Visiting Scholar at Melbourne Law School in late 2011, participating in the Obligations Group and working on his forthcoming book Legislation and Reform in the Law of Obligations. At Birmingham, James teaches Tort, Equity and Trusts, Advanced Torts and Advanced Property, and he has collaborated on teaching innovations with Dr Katy Barnett of Melbourne Law School. James has recently been appointed as one of the co-editors for the 20th edition of the leading Trusts book, Hanbury and Martin’s Modern Equity (to be published by Sweet & Maxwell in 2015). James presented on some of the issues in this post at the Fifth Biennial Conference on the Law of Obligations in 2010, which was hosted at St Anne’s College, Oxford, by the University of Melbourne and the University of Oxford. In 2014, James will be presenting at the Seventh Biennial Conference, which is being jointly organised by the University of Melbourne and the University of Hong Kong.

Those interested in further reading on this theme and more broadly on precedent may wish to read:

  • Andrew Burrows, ‘The Australian Law of Restitution: Has the High Court Lost Its Way?’ in Elise Bant and Matthew Harding (eds), Exploring Private Law (Cambridge University Press, 2010) 67.
  • Chief Justice R S French ‘State of the Australian Judicature’ (2010) 84 Australian Law Journal 310.
  • James Lee, ‘Fidelity in Interpretation: Lord Hoffmann and the Adventure of the Empty House’ (2008) 28 Legal Studies 1.
  • Matthew Harding and Ian Malkin, ‘Overruling in the High Court of Australia in Common Law Cases’ (2010) 34 Melbourne University Law Review 518.
  • Matthew Harding and Ian Malkin, ‘The High Court of Australia’s Obiter Dicta and Decision-Making in Lower Courts’ (2012) 34 Sydney Law Review 239.
  • R McColl, ‘The Art of Judging’ (2008) 12 Southern Cross University Law Review 43.
  • Norman O’Bryan and Chris Young ‘A View from outside the Vortex on Keith Mason’s Retirement Speech and the Australian Doctrine of Judicial Precedent’ (2008) 82 Australian Law Journal 771.
  • James Lee (ed), From House of Lords to Supreme Court: Judges, Jurists and the Process of Judging (Hart Publishing, 2011), which includes an essay by the Hon Michael Kirby AC CMG: ‘From Appellate Committee to the UK Supreme Court: Independence, Activism and Transparency’.

7 thoughts on “Precedent on High: The High Court of Australia and ‘Seriously Considered Dicta’

  1. James, thanks for your post, which with I thoroughly agree. I thought I should mention a possible variation of the new ‘Australian way’ that also emerged from the High Court roughly contemporaneously with Farah.

    It is a case from the criminal law, Philips v R [2006] HCA 4. The bulk of that case involved a (controversial) ruling about the ‘relevance’ in a rape trial of a separate accusations against the accused of rape by other complainants, in circumstances where the accused claims that all the incidents were consensual . However, in (presumably seriously considered) dicta, the Court also noted criticisms of a broader rule about so-called ‘similar fact evidence’ that the Court had announced in a 1995 judgment, Pfennig v R [1995] HCA 7. The Court’s concern wasn’t to respond to those criticisms (which didn’t need to be addressed to resolve the appeal before it), but rather to address the role of intermediate courts in applying criticised High Court judgments.

    At [60], the Court set down a general approach that I think everyone would agree with:

    “It must be said at once that it is for this Court alone to determine whether one of its previous decisions is to be departed from or overruled. Of course, in criminal cases it is often necessary for trial judges and Courts of Criminal Appeal to elaborate upon rulings of this Court; to gather together rules expressed in several cases; to apply rules to different facts; and sometimes to reconsider rules affected by later legislation. Within spaces left by the binding determinations of this Court, trial judges and intermediate courts retain their proper functions. However, these do not extend to varying, qualifying or ignoring a rule established by a decision of this Court. Such a rule is binding on all courts and judges in the Australian Judicature.”

    However, the Court went on to say (at [64]) that, because an earlier Queensland court had, after an extensive analysis of the 1995 judgment, set down a test for similar fact evidence that was ‘expressed differently’ from the High Court’s own test, the Queensland Court’s test ‘should not be adopted or applied’. What is noteworthy is that the Court reached this conclusion without any express analysis (or even a description) of the Queensland test. It was sufficient that ‘it cannot be assumed that in every case’ a differently expressed test ‘would operate identically to the tests expressed’ by the High Court. My view is that that may leave an especially small ‘space left by the binding determinations of’ the High Court when it comes to intermediate courts interpreting the reasons of the Court.

    (The Court did additionally assert that the Queensland Court’s reasons ‘might be read as suggesting’ that a different operation was intended by the intermediate court. Obviously, if that claim was correct, that would be important, but that certainly isn’t my reading of the lower court’s reasons and the High Court doesn’t cite what part of the reasons it refers to. It may be that the High Court’s reading is based on the undoubted criticisms the Queensland court made of the earlier judgment, but if such criticisms were sufficient to dispose of any test that intermediate court developed to explain how a High Court rule operates, then that would also be an unfortunate restriction on the role of intermediate courts. A particular reason to doubt the High Court’s reading is that, in Phillips’s own case, both the prosecution and the defence consented to the use of the Queensland court’s test, something that surely would not have happened if it was thought to operate differently to the High Court’s test.)

  2. This is a really interesting post, James. It is worth noting, however, that following obiter dicta of a higher court (or, by extension, considering themselves bound by such dicta) does not always fetter the lower courts in their development of the law. Such dicta could in fact effect a liberalisation where they doubt a previously restrictive approach. An example of this phenomenon is the debate over whether the lower courts in England and Wales should follow Lloyds v Rosset or pick up on dicta of the House of Lords/Supreme Court in Stack v Dowden and Jones v Kernott in the context of the context of the common intention constructive trust (see my paper at

  3. In an Australian context, Farah has been seen as an admonition to superior State courts (in particular the New South Wales Court of Appeal) against liberalising doctrines. But you are right – following dicta of superior courts can sometimes lead to an extension of the law rather than a restriction. It would be rather ironic if a liberalising dicta were picked up and a State court expanded the law as a result of Farah! (The law of unintended consequences maybe?)

  4. James, I wonder whether you are reading too much into paragraphs 134 and 158 of Farah. On my reading, the Court is critical principally of the adventurism of the NSW Court of Appeal in reshaping the first limb of Barnes v Addy. The NSWCA’s approach was worrying not only for its lack of basis in authority but for its misunderstanding of the High Court’s commentary by way of dicta. I do not read paragraph 134 of the reasons as demanding adherence to dicta in the same way as adherence to ratio is required.

    If I’m wrong about that, then one has to ask this question: by what right does the High Court require adherence to anything other than the rationes of its decisions? Even accepting that the final court of appeal has (at a practical level) a legislative function, there is no room for that role to extend to modifying the doctrine of precedent.

    Any judge worth his/her salt would have to reject any submission that he/she is bound to do more than accord dicta in the High Court the respect which it obviously deserves.

  5. Hi Nick, my own view is set out in my comment on Matthew Harding’s earlier post. I personally believe that the dicta in Consul were unclear. I have seen vastly differing interpretations of the level of knowledge that Gibbs J in Consul Developments required to make out the first limb. Moreover, I saw Gibbs J in Consul as reading down the “dishonest and fraudulent design” requirement, but the High Court doesn’t mention this in Farah when it discusses dishonest and fraudulent design (it just focuses on Stephen J). The difficulties with requiring the trustee to have a dishonest and fraudulent design were also noted by Lord Nicholls in Royal Brunei v Tan.

    That being said, I think the NSWCA could have tackled the matter of restitutionary liability in a better way (as I note in my other comment).

    The High Court may simply have meant to chastise the NSWCA with regard to developments on the particular doctrine in question, but I agree with James that it has been understood much more broadly than that by state courts. In my opinion, it has had a chilling effect on the development of doctrine by state courts (whether this was intended or unintended).

  6. Thank you for that, Nick and Katy. I agree with Katy, really: it would in context be perfectly possible to understand the thrust of Farah as a specific rebuke on a specific point (Mason P’s perhaps somewhat mischievous adventurism, as you say, and to which I advert in the final paragraph). But the tone seems broader than that, and the subsequent interpretation in cases, at both State and HCA level, seems to be wider. And, Heydon J’s article indicates the same extent. And it may not be the same level of adherence as required for the ratio, but it is clearly a stronger level of adherence than might otherwise be thought.
    And, again, either way: it behoves the Court to explain.

Comments are closed.